How to Use Microbiome Signatures as Biomarkers
Gwynneth Ballentine PhD; Jean-Philippe Laine PhD; Hajo Schiewe PhD
EXECUTIVE SUMMARY
Validated molecular biomarkers have become an increasingly important part of developing and
prescribing drugs more efficiently as well as diagnosing physiological states earlier or more reliably.
Enabled by improvements in next generation sequencing technologies, the microbiome (the entire
collection of bacteria, viruses, and fungi that naturally occupy various niches on or in a given
environment) has become a strategic source for biomarker discovery in the drug development and
diagnostic industry. In this paper, we provide an introduction to microbiome-based biomarkers and the
use of microbiome data as one component of a larger data profile for individuals in this era of precision
medicine. We discuss the application of biomarkers enabling “targeted therapies” or serving as surrogate
efficacy markers to shorten otherwise lengthy clinical trials. We explore the role biomarkers play in
reimbursement decisions for new therapies based on cost-effectiveness1. Finally, we provide an example
process for biomarker identification and validation. Details regarding Diversigen’s unique approach and
expertise are provided for consideration.
THE CHALLENGE
The latest analysis of the average cost to develop and win marketing approval for a new drug has been
estimated as being higher than 2.5 billion USD with an approximate even split between out-of-pocket
costs and time costs2. Innovative approaches are needed to increase efficiency in the drug development
process in order to get new, more efficacious treatments to patients more quickly and at lower costs.
The stratification of both healthy individuals and
patients through the use of validated biomarkers is an
integral part of precision medicine. The application of
these biomarkers, enabling “targeted therapies”, range
from predicting drug response, adverse effects, disease
predisposition and disease progression as well as
serving as surrogate efficacy markers to shorten
otherwise lengthy clinical trials. To date, the majority of
pharmacogenomic biomarkers for various types of
therapies are “host” mutations/polymorphisms in either
drug target receptor genes, e.g. EGFR, or drug
metabolism enzyme genes like CYPs, but other factors
such as the microbiome community structure have to be
taken into account when developing a targeted therapy.
Besides improving the standard of care and increasing
patient benefit, biomarkers play an increasingly
important role when it comes to reimbursement
decisions for new therapies based on costeffectiveness1. Of note, the ten highest-grossing drugs
currently in the market have a patient benefit ratio of
25%, at best, to an appallingly low ratio of 4% (Figure
1)3. Interestingly, the target indications for many of the
highest-grossing drugs have all been recently
Fig. 1: Drug Response Rate for the ten highest grossing drugs3
linked to dysbiosis in the microbiome4,5,6.
1
THE MICROBIOME – AN EMERGING FIELD FOR BIOMARKERS
The field of microbiome research has grown exponentially over the past several years, catalyzed by the
US-led Human Microbiome Project (HMP) 7 and the European-led MetaHIT 8 project. Study design and
methods are evolving from strictly descriptive approaches to those that include comprehensive functional
analysis of relevant microbial communities and constituents9. Microbial cells in the human body not only
outnumber human cells by approximately ten to one1 but their genetic load has been estimated between
100 to 200:1 greater than the human genome. So their influence on human development as well as
disease pathogenesis and progression is a significant area of interest 10 . Countless efforts are now
underway to develop microbiome-based diagnostics, therapeutics, and services in both the academic and
commercial arenas.
Indeed, a healthy symbiotic microbiome has now been shown to have a direct role in the:

Establishment of innate immunity

Host metabolism (Energy, Vitamins, Digestion)

Drug metabolism

Production of signaling molecules with local or distant targets
Therefore, one area of marked focus is the ability to use changes in microbial community signatures, e.g.
dysbiosis, to interrogate potential biomarkers associated with risk, onset or exacerbation of disease11 as
well as for prediction of treatment success.
Several mechanisms of action associated with a dysbiotic microbial community have been implicated in
many disease states, such as:

Disruption of microbial community homeostasis (e.g. C.diff.12,13,14, H.pylori, viral upper
respiratory infections15,16)

Interference with immune regulation and maturation17 (e.g. diabetes18, IgA secretion18, T cell
over-reactivity19)

Stimulation of host-gene expression (e.g. cancer20,21)

Shift in nutrient metabolism (e.g. malnutrition22, obesity23)
One of the most highly studied indications is Clostridium difficile infections, in which microbial
dysbiosis has been directly linked to onset as well as chronic recurrence. FMT treatment has become
more common, but its outcome/success seems under developed. 12,13,14.
In a model of Type 1 diabetes (T1D), there is evidence that certain shifts in the microbial signature can
be linked to both onset and progression of the disease13. To better understand this link in humans and
establish a microbiome causality in the development of T1D, a sub-study was established in the multinational endeavor, TEDDY (The Environmental Determinant of Type 1 Diabetes in the Young), to analyze
over 22,000 GI, plasma, and nasal swab samples from birth to disease onset. Dr. Joseph Petrosino, Chief
Science Officer at Diversigen, is spearheading the microbiome component of this NIH-funded project, the
largest clinical microbiome study to-date.
For various forms of cancer20, enrichment or augmentation of specific bacteria compared to a control as
well as links with other clinical endpoints, including BMI and gene profile changes, are paving the way for
better preventative and diagnostic tools21.
Additionally, a gut microbial profile linked to obesity has been shown to be transmissible, unveiling a
unique way to identify risk markers of the disease23.
2
Studies indicate that mitochondrial dysfunction impairs synaptic function and plays an important role in
neurodegenerative disease pathology24. Given that certain microbial community structures result in an
overproduction of propionic acid, which – when metabolized to methyl-malonyl-CoA – negatively impacts
mitochondrial efficiency25, stratification of subjects for clinical trials in the neurodegenerative field based
on such community structures has the potential to significantly reduce cost and time.
Environmental factors, such as diet and prolonged antibiotic use are suggested triggers of microbial
dysbiosis and other microbial shifts that may cause an increase in disease risk for, and/or an earlier onset
and more aggressive progression of disease11-13,17. Linking the human commensal microbiota and the
mechanisms and triggers underlying disease, in the context of biomarker identification, is quickly
becoming the most clinically relevant use of microbiome data. This information, which may be one
component of a larger data profile for individuals in this era of personalized or precision medicine, may be
used for risk prediction and/or diagnosis of disease or disease progression and may also be used for
stratification of subjects in clinical drug trials.
A CUSTOMIZED APPROACH FOR BIOMARKER PROJECTS
At Diversigen, we utilize our extensive knowledge base, flexibility in sequencing platforms, and an
agnostic approach to analysis to identify the appropriate method for microbial community or metabolic
profiling for each project. With an emphasis on reaching adequate sequencing depth and delivery of high
quality data, we are able to infer metabolic potential for more targeted hypothesis generation under
which specific biomarkers could be identified and validated.
We offer approaches to address every level of microbiome query. We begin with amplicon-based
approaches, such as 16S rDNA sequencing, which yield a high-level overview of the structure of a
microbial community. We then have the flexibility to utilize multiple different strategies, including whole
genome shotgun (WGS) DNA and/or RNA sequencing to profile the species in a microbial community, as
well as map the functional and metabolic potential of the microbes in a particular sample. When a strain
of interest is isolated, we then can use long read sequencing platforms and library construction strategies
to assemble accurate, complete bacterial genomes.
Fig 2: Example process for identifying, validating and developing microbiome-based biomarkers
3
Our customers benefit from our experience in both human clinical trials and animal model studies of
numerous human diseases, as well as our experience linking other ‘omics data with
microbiome/metagenomic data. For example, our expertise enables us to link unique metabolite
signatures to various bacteria of interest based on metabolomics data.
The value of microbiome-based biomarkers is clear when considering that the majority of samples taken
for microbiome analysis can be either non- or minimally invasive without the need for biopsies. This will
help with increased patient compliance when it comes to testing for various diseases.
Fig. 3: Sampling strategies for microbiome-based biomarker discovery and validation
CONCLUSION
The future looks bright as research efforts continue to utilize microbiome data for biomarker discovery
across multiple disease states. Potential impact of this research includes the development of novel
approaches associated to diagnose risk, onset, or exacerbation of disease resulting in significantly
increased patient benefit ratios for successful drugs. Microbiome information, which may be one
component of a larger data profile for individuals in this era of personalized medicine, may be used for
risk prediction and/or diagnosis of disease or disease progression and for stratification of subjects in
clinical drug trials.
Key factors in successful microbiome-based biomarker identification and validation include stringent
sampling protocols, high quality metagenomic data with sequencing depth adequate to identify all key
players, and the ability to identify bacteria of interest. When properly employed, these factors can also
result in the successful linking of metabolic and functional profiles to human or animal phenotypic data.
Diversigen offers extensive experience, a full-service lab, and a customized approach to maximize the
potential of biomarker projects. For more information, contact the authors of this paper at
[email protected].
ABOUT DIVERSIGEN
With years of experience, flexible sequencing, and expert bioinformatic capacity, Diversigen is building on
research conducted at Baylor College of Medicine in the Alkek Center for Metagenomics and Microbiome
Research. Our highly qualified and experienced personnel are recognized as thought leaders at the
forefront of the microbiome field. Diversigen provides comprehensive services that include a variety of
4
amplicon-based and metagenomic sequencing, expert bioinformatics with timely turnaround of high
quality sequencing data, in-depth data analysis, customized project design, professional project
management and customer-focused service. Diversigen’s lab is soon be CLIA-certified.
References
1
McCabe C et al.; The NICE cost-effectiveness threshold: what it is and what it means; Pharmacoecon, 2008; 26(9):733-44
2
DiMasi J; Innovation in the Pharmaceutical Industry: New Estimates of R&D Costs;
http://csdd.tufts.edu/files/uploads/Tufts_CSDD_briefing_on_RD_cost_study_-_Nov_18,_2014..pdf; accessed on Sept. 18th, 2015.
3
Schork J; Personalized Medicine: Time for one-person trials; Nature, 2015, 520:609-611
4
Huang YJ, Boushey, HA; The microbiome in asthma, J Allergy Clin Immunol, 2015, 135(1):25-30
5
Tinan TG et al.; Genomics of schizophrenia: time to consider the gut microbiome?, Molecular Psychiatry, 2014, 19:1252-57
6
Cho I, Blaser MJ; Human Microbiome: at the interface of health and disease, Nat Rev Genet., 2012, 13(4):260-270
7
s. www.hmpacc.org
8
Qin J et al.: A human gut microbial gene catalogue established by metagenomic sequencing; Nature, 2010, 464:59-65
9
The iHMP Network Consortium; The Integrative Human Microbiome Project: Dynamic Analysis of Microbiome-Host Omics Profiles
during Periods of Human Health and Disease; Cell Host & Microbe; 2015, 16: 276-289
10
Gross L; Human Gut Hosts a Dynamically Evolving Microbial Ecosystem, PLoS Biol; 2007, 5(7): e199
11
Dietart RR at al.; Biomarkers for the 21st century: listening to the microbiome; Toxicol Sci, 2015, 144(2): 208-216
12
Seekatz AM et al; Clostridium difficile and the microbiota; J Clin Invest; 2014, 124(10: 4182-4189
13
Koenigsknecht MJ et al.; Faecal microbiota transplantation for the treatment of recurrent Clostridium difficile infection: current
promise and future needs; Curr Opin Gastroenterol, 2013, 29(6): 628-632
14
Taur Y, Pamer EG; Harnessing microbiota to kill a pathogen: Fixing the microbiota to treat Clostridium difficile infections; Nat
Med, 2014, 20(3): 246-247
15
Allen EK et al.; Characterization of the nasopharyngeal microbiota in health and during rhinovirus challenge, Microbiome, 2014,
25;2:22
Yi H et al.; Profiling bacterial community in upper respiratory tracts, BMC Infect Dis., 2014, 14:583
Kau AL et al; Human nutrition, the gut microbiome and immune system: envisioning the future; Nature, 2012, 474(7351): 327336
16
17
18
Brown K et al.; Prolonged antibiotic treatment induces a diabetogenic intestinal microbiome that accelerate diabetes in NOD mice;
ISME J, 2015, 10: 114
19
Yu B et al.; Dysbiosis of gut microbiota induced the disorder of helper T cells in influenza virus-infected mice, Hum Vaccin
Immunother., 2015, 11(5):1140-6
Narayanan V et al; Human Fecal Microbiome-Based Biomarkers for Colorectal Cancer; AACR J, 7(11): 1108-1111
20
21
Gui QF et al; Well-balanced commensal microbiota contributes to anti-cancer responses in a lung cancer mouse model; Genet Mol
Res; 2015, 14(2): 5642-5651
22
Preidis GA et al.; Composition and function of the undernourished neonatal mouse intestinal microbiome, J Nutr Biochem., 2015,
26(10):1050-7
Turnbaugh PJ et al; An obesity-associated gut microbiome with increased capacity for energy harvest; Nature, 2006, 444: 10271031
23
24
D. Pathak et al.; The Role of Mitochondrially Derived ATP in Synaptic Vesicle Recycling; J Biol Chem; 2015 290(37): 22325-22336
25
R. Frye et al.; Gastrointestinal Dysfunction in Autism Spectrum Disorder: The Role of the Mitochondria and the Enteric
Microbiome; Microb Ecol Health Disease; 2015, 26: 27458-74
5